# sep.sys: 2-stage population sensitivity In RSurveillance: Design and Analysis of Disease Surveillance Activities

## Description

Calculates population-level (system) sensitivity for representative 2-stage sampling (sampling of clusters and units within clusters), assuming imperfect test sensitivity and perfect test specificity

## Usage

 `1` ```sep.sys(H = NA, N = NA, n, pstar.c, pstar.u, se = 1) ```

## Arguments

 `H` population size = number of clusters in the population, default = NA `N` population size within clusters, scalar or a vector of same length as n, default = NA `n` sample size (vector of number tested per cluster) `pstar.c` cluster (herd) level design prevalence, scalar, either proportion or integer `pstar.u` unit (animal) level design prevalence, scalar, either proportion or integer `se` unit sensitivity of test (proportion), scalar, default = 1

## Value

list of 6 elements, 1) population level sensitivity, 2) vector of cluster-level sensitivities, 3) N, 4) n, 5) vector of design prevalences and 6) unit sensitivity

## Note

if pstar.c is not a proportion N must be provided (and N>=n)

## Examples

 ``` 1 2 3 4 5 6 7 8 9 10 11 12 13 14``` ```# examples for sep.sys - checked H<- 500 N<- rep(1000, 150) N[5]<- NA n<- rep(30, 150) pstar.u<- 0.1 pstar.c<- 0.01 se<- 0.98 sep.sys(H, N, n, pstar.c, pstar.u, se) sep.sys(NA, N, n, 0.02, 0.05, 0.95) N<- round(runif(105)*900+100) n<- round(runif(105)*30+10) sse<- sep.sys(1000, N, n, 0.02, 0.05, 0.9) data.frame(N, n, sse[[2]]) ```

RSurveillance documentation built on May 29, 2017, 11:52 p.m.